evals-first

Enforce write-first evaluations for LLM-interacting code using Claude Code sessions.

Updated Apr 13, 2026

Search Tags:#code review #skill validation #prompt testing #eval-first #LLM-interaction

Ready to run

npx skills add https://github.com/youssefKadaouiAbbassi/yka-code --skill evals-first

💬 Prompt for ChatGPT / Claude Chat▼

Please help me install this Agent Skill.
Skill: evals-first
Source: https://github.com/youssefKadaouiAbbassi/yka-code/tree/main/skills/evals-first
Command: npx skills add https://github.com/youssefKadaouiAbbassi/yka-code --skill evals-first

SYSTEM DOCUMENTATION & REQUIREMENTS

💡 This Skill includes scripts (resource) components.

What problem does it solve?

The 'evals-first' skill addresses the challenge of ensuring the correctness of prompt and skill logic in code that interacts with large language models (LLMs) before the actual implementation is made.

Core Features & Use Cases

Eval-First Process: Ensures writing and testing evaluations before implementing the actual logic.
LLM Interaction: Specifically designed for code that interacts with Claude Code sessions.
Regression Coverage: Provides LLM-behavioral regression coverage, complementing traditional unit testing.
Use Case: When a new feature is proposed that affects prompt logic or skill routing, this skill is activated to create and verify the evaluation before any implementation changes are made.

Quick Start

Load the 'evals-first' skill and write an evaluation that fails before you modify the prompt logic or skill in your code.

Frequently Asked Questions about evals-first

High-intent search queries and answers about installing and using this skill.

FAQPage Schema

How do I test prompt logic before implementing LLM code changes?▼

To test prompt logic before implementation, you enforce an eval-first process by writing an evaluation that fails before modifying any prompt logic or skill routing in your LLM-interacting code.

What is eval-first discipline for LLM-interacting code?▼

Eval-first discipline is a development process that requires writing and verifying behavioral evaluations for LLM-interacting code before the actual implementation is made, ensuring prompt and skill logic correctness.

Can I use this to get behavioral regression coverage for Claude Code sessions?▼

Yes, this process provides LLM-behavioral regression coverage specifically designed for code that interacts with Claude Code sessions, complementing traditional unit testing.

When do I need to write evaluations for prompt logic?▼

You need to write evaluations for prompt logic when a new feature is proposed that affects prompt logic or skill routing, activating this process to verify the evaluation before implementation changes.

Does eval-first testing replace traditional unit testing?▼

Eval-first testing does not replace traditional unit testing; it complements it by providing LLM-behavioral regression coverage to ensure the correctness of prompt and skill logic.

How do I start with prompt testing for skill validation?▼

To start prompt testing for skill validation, load the evals-first skill and write an evaluation that intentionally fails before you modify the prompt logic or skill in your code.